Sponsored by the Delta Science Program and the UC Davis Muir Institute

RESEARCH Predator Diet Analysis: A Comparative Study Dylan K. Stompe*1,2, Jason D. Roberts2, Carlos A. Estrada1, David M. Keller1, Nicholas M. Balfour1, and Amanda I. Banet1

ABSTRACT similar as determined through index of This study examined diets of two predatory relative importance and PERMANOVA fish species, the native Sacramento modeling. Water temperature was the only Pikeminnow ( grandis) and the variable that significantly affected diet introduced Striped Bass (Morone saxatilis), in composition. Results reflect similar dietary the Sacramento River, , USA. Both niches for both species in the Sacramento species have been implicated in native species River. declines through predation, eliciting our investigation of their diets in the Sacramento River. Sampling occurred between March KEY WORDS and November 2017, and was conducted via Sacramento Pikeminnow, predation, Striped hook and line on a 35-km reach near Chico, Bass, introduced species, California, fisheries, California. Habitat types sampled include water diversion, Chinook Salmon engineered structures (water diversions and beam bridges), rip-rapped channel edges, and natural riverbank. Stomach contents INTRODUCTION were collected via gastric lavage and later Because predation is a challenge with which processed using visual, gravimetric, and nearly all organisms must contend, it is often genetic techniques. Diets of Sacramento considered in the management of vulnerable Pikeminnow and Striped Bass were highly populations of fishes (Zimmerman and Ward 1999; Link 2002). In the Sacramento River, California, this challenge may be amplified SFEWS Volume 18 | Issue 1 | Article 4 for populations of juvenile Chinook Salmon https://doi.org/10.15447/sfews.2016v18iss1art4 by climate change, a complex water diversion system, hatchery domestication effects, lack * Corresponding author: [email protected] of juvenile rearing habitat, and non-native 1 Department of Biological Sciences California State University, Chico predatory fish species (Brown and Moyle Chico, CA 95929 USA 1981; NOAA Fisheries 2018). Factors such as 2 Northern Region climate change and habitat loss have served California Department of Fish and Wildlife to limit the natural production potential of Redding, CA 96001 USA Chinook Salmon (Oncorhynchus tshawytscha)

1 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

in the Sacramento River, while domestication Pikeminnow and Striped Bass are common of hatchery-reared fish and watershed in the Sacramento River, despite overall engineering increase the vulnerability of population declines throughout the watershed juvenile Chinook Salmon to predation by (Stevens et al. 1985; Kohlhorst 1999; Moyle both native and non-native predators (NOAA 2002; Brown and Moyle 2005). Fisheries 2018). Sacramento Pikeminnow and Striped Bass Of the four Chinook Salmon runs native have both been considered as potential to the Sacramento River, winter-run are contributors via predation to native species listed as endangered, spring-run are listed decline in the highly modified Sacramento as threatened, and both Central Valley fall- River system (CDFG 1999; Moyle 2002; run and late fall-run have been identified Lindley and Mohr 2003; Bonham 2011). as California species of special concern Pikeminnow species (Ptychocheilus spp.) have (Moyle et al. 2015; NOAA Fisheries 2018). been shown to consume large amounts of In addition, Green Sturgeon (Acipenser outmigrating salmonids from a range of US medirostris) and Steelhead (Oncorhynchus West Coast watersheds under hydraulically mykiss), both species native to the Sacramento favorable conditions, and, in response, River, are also listed as threatened under predator control measures have been the federal Endangered Species Act. Prior implemented with mixed results (Brown studies have investigated the potential for and Moyle 1981; Brown and Moyle 1997; population-scale effects of predation by non- Tucker et al. 1998; Friesen and Ward 1999; native species on prey populations, including Zimmerman and Ward 1999; Moyle 2002; listed Chinook Salmon, through population Nakamoto and Harvey 2003). Likewise, modeling (Lindley and Mohr 2003), Striped Bass have also been shown to bioenergetic modeling (Loboschefsky et al. consume salmonids (Stevens 1966; Tucker 2012), and telemetry survival studies (Cavallo et al. 1998), which modeling suggests may et al. 2013) within the Sacramento River and have population-scale effects (Lindley and adjacent watersheds. While these studies Mohr 2003; Sabal et al. 2016). Within the identify predation as a tangible “top-down” Sacramento River drainage, the effects of control of prey populations, population- dams and diversions on predation have been scale effects are often difficult to quantify a point of concern (Brown and Moyle 1981; in the highly dynamic and heterogeneous Tucker et al. 1998; Sabal et al. 2016). Altered Sacramento River system. water dynamics near these structures may serve as ambush habitat, making conditions The middle Sacramento River contains favorable to opportunistic predators such as two abundant piscivorous fish species: Striped Bass and Sacramento Pikeminnow. Sacramento Pikeminnow (Ptychocheilus In addition, juvenile Chinook Salmon show grandis) and Striped Bass (Morone saxatilis). In a preference for low-water-velocity areas the Sacramento River and other California (Hillman et al. 1987), and may be more watersheds, both predators consume susceptible to predation when exposed to vulnerable native species, including juvenile disorienting flows caused by engineered Chinook Salmon and Steelhead (Stevens structures (Brown and Moyle 1981; Tucker et 1966; Thomas 1967; Brown and Moyle 1981; al. 1998; Deng et al. 2010; Sabal et al. 2016). Brown and Moyle 1997; Nakamoto and Harvey 2003; Sabal et al. 2016). Sacramento Striped Bass long avoided management as a Pikeminnow are native to the Sacramento predatory species because of their value as River drainage; Striped Bass were introduced a game fish; that changed, however, in the in 1879 as a recreational and commercial 1990s, when the California Department of species (Moyle 2002). Both Sacramento Fish and Game (now California Department

2 MARCH 2020

of Fish and Wildlife; CDFW) ceased stocking Piscivory in Striped Bass generally begins at Striped Bass so as not to facilitate predation around 15 cm total length (Texas Instruments on listed salmonids (CDFG 1999; Moyle 1974) as determined by gape limitation; 2002). Several studies investigated diets of however, piscivory has been shown to have Striped Bass in the Sacramento−San Joaquin a strong seasonal association in the Delta Delta (Delta); however, there is little recent (Nobriga and Feyrer 2007). Males become literature on Striped Bass diets in the middle sexually mature at 2 years of age; females reaches of the Sacramento River below generally do not mature before 4 years of age Shasta Dam (Brown and Moyle 1981; Tucker (Moyle 2002). Immature individuals may be et al. 1998; Moyle 2002; Nobriga and Feyrer found throughout the system, while mature 2007). Likewise, although Nobriga and Feyrer individuals exhibit an anadromous life- (2007) compared the diets of Sacramento history strategy, spending much of the year Pikeminnow and Striped Bass concurrently in the San Francisco Estuary (Sacramento− in the Delta, diets for these two species have San Joaquin Delta, Suisun Bay and Marsh, not been compared in the middle Sacramento San Pablo Bay, San Francisco Bay) and the River since Tucker et al. (1998) investigated Pacific Ocean (Thomas 1967; Moyle 2002). predation around the Red Bluff Diversion Mature Striped Bass generally enter the Dam (RBDD) in the mid-1990s. The middle Sacramento River system in spring to spawn, Sacramento River is colder and is generally before returning to downstream habitats a single channel, as opposed to the Delta’s (Moyle 2002); however, resident contingents warmer network of channels or open water. of mature adults exist in their native US East It contains a different assemblage of prey Coast rivers (Morris et al. 2003), and recent species (Moyle 2002), some of which are only studies on California Striped Bass indicate seasonally available, potentially affecting there may be both resident and migratory predator diets. We investigated the diets of contingents within the Sacramento River these two species over a year, using visual watershed, as well (Le Doux−Bloom 2012; and genetic techniques to identify prey Sabal et al. 2019). species. Using these methods, we describe Sacramento Pikeminnow and Striped Bass Study Reach diets by quantifying: (1) relative importance Fish were collected on the Sacramento River, of prey, (2) overlap of predator diets, and (3) between Ord Bend boat ramp (river kilometer effects of engineered structures on predator 296) and Glenn−Colusa Irrigation District diet. diversion facility (GCID; river kilometer 331; Figure 1). We chose this logistically manageable reach because of the presence MATERIALS AND METHODS of both Striped Bass and Sacramento Study Organisms Pikeminnow, seasonal populations of Sacramento Pikeminnow are native to the migratory salmonids, engineered structures, Sacramento River and are an abundant and its diversity of habitat. piscivorous fish within the freshwater portions of the system. They become We implemented a balanced sampling piscivorous at 10 to 20 cm total length, design, breaking the total sampling reach sexually mature at 3 to 4 years of age, and into four sections of similar length, each of may exhibit either resident or migratory which contained three fixed sampling sites life-history strategies within the fresh and of different habitat types. Habitat types brackish portions of the system (Moyle 2002; included engineered, rip-rap, and natural Nobriga et al. 2006). locations. Engineered sites were adjacent to engineered structures such as water diversion facilities or beam bridges. Rip-rap sites had

https://doi.org/10.15447/sfews.2020v18iss1art4 3 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

per sampling period, with all three habitat GCID types within a section sampled for a total of Section 1 3.75 hours. Individual sections were sampled biweekly, and alternated between morning and evening. Before sampling a section, we randomly generated the order in which we visited sites. Section 2 Hook and line sampling began in March 2017 and continued through the end of November 2017. High water and unsafe conditions prevented sampling in January Section 3 and February, as well as part of March and April, and limited crew availability was responsible for lack of sampling in December. riprap We sampled from a 4.9-m jet boat at anchor. natural Four rods were fished in randomized order, Section 4 manmade each of which was assigned a unique bait Ord Bend (large sardine piece, small sardine piece, 4 km chicken liver, and nightcrawlers) that would not contaminate subsequent diet sample Figure 1 Location of hook and line sampling sites by site analysis. We selected bait types to attract type and section. Map courtesy of Dr. Paolo Segre, Stanford target species from a range of sizes. When University. we caught fish, we removed them from the water, measured them for length (fork length; cm) and weight (kg, ± 50 g), and at least one adjacent bank that had been placed them into an aerated holding tank. armored with large rock. Natural sites were We then pumped fish stomachs using non- not near either armored bank or engineered lethal gastric lavage, a method in which structures. Since the limiting habitat type pulses of pressurized water are directed into in our study reach was engineered habitat, the esophagus, causing the fish to evacuate the number of rip-rap and natural sites its stomach contents (Foster 1977). To retain corresponded to the number of engineered sample integrity after returning from the sites available. We identified possible rip-rap field, stomach content samples were collected and natural sites, and randomly selected a in a fine mesh bait net (flushed with river subset of four of each site type. water between samples), transferred to sterile Whirl-pak® bags, labeled, stored on ice, and Sample Collection frozen at -20 °C immediately. We used hook and line sampling methods to collect data from wild Sacramento We measured water temperature at each Pikeminnow and Striped Bass. We selected site via the transducer from the onboard hook and line sampling because of its low fish-finder/GPS unit (Garmin® Striker 4), material cost and low risk of listed species and measured water clarity with a bycatch. We generally sampled twice weekly, 20.3-cm-diameter white and black Secchi with one morning shift (starting 45 minutes disk. Upon two occasions, when crew failed before sunrise) and one evening shift to record water clarity, it was estimated (starting 5 hours before sunset) to control as the mean of measures taken during the for diurnal feeding effects (Fraser et al. previous and following sampling days. 1993). We sampled each site for 1.25 hours

4 MARCH 2020

Laboratory Analysis cannot be easily identified—a process that Stomach contents were processed to accelerates with increasing water temperature determine weight (g, ± 0.001g), frequency (Vondracek 1987). Genetic methods, although of occurrence, and number of prey within more costly and labor-intensive than simple each sample. Stomach content samples visual identification, allowed for a more were removed from the freezer and allowed holistic representation of predator diets to fully thaw at room temperature. Using (Valdez−Moreno et al. 2012). instruments sterilized in 20% bleach solution, we placed samples onto new We chose quantitative PCR (qPCR) as the polystyrene weigh boats, sorted them, and primary analysis technique because of its segregated prey by taxonomic category. higher throughput than traditional PCR, Prey categories included fish, crayfish, other as well as its ability to quantify sample macroinvertebrates (primarily terrestrial DNA, allowing for discrimination of true and aquatic insects), and unknown soft versus contaminant DNA (Rees et al. 2014). matrix (i.e., visually unidentifiable yet Although still a relatively novel method for clearly organic stomach content material; identifying prey from stomach contents, what remains after less-digested prey is qPCR has proven successful for this removed). We used diagnostic parts—such application in a number of studies (Durbin as spinal columns for fish, and head parts et al. 2011; Hunter et al. 2012; Taguchi et al. for macroinvertebrates—to enumerate 2014; Michel et al. 2018). individuals. We determined a reference list of potential Next, we weighed the prey group to the prey species through the lead author’s nearest thousandth of a gram using an previous CDFW seining and snorkel survey Ohaus® STX223 Scout portable balance, after experience, and verified them using Moyle blotting prey dry with new paper towels. (2002). We referenced previously-designed For individual fish that could be clearly prey species primers from current literature identified as such, we removed a sample of (Jordan et al. 2010; Baerwald et al. 2012; tissue (≤ 0.25g) and transferred it to a new Brandl et al. 2015), and designed additional 1.5-mL micro-centrifuge tube, labeled it, and primers using Genbank sequence data stored it on ice. When individuals could not and the National Center for Biotechnology be clearly differentiated, we homogenized the Information (NCBI) Primer-BLAST tool soft-matrix material with dissecting tools, (Benson et al. 2012; Ye et al. 2012). We and took at least one representative grab tested primer sets for validity against sample (≤ 0.25g). Once a processing session known voucher tissue via PCR and qPCR, was completed, resulting pre-category sub- and optimized them to determine correct samples were immediately returned to storage annealing temperatures. Testing and optimizing at -20 °C for later genetic analysis. was done for all primer sets except for River Lamprey (Lampetra ayresii) and Western Brook Genetic Analysis Lamprey (Lampetra richardsoni), as the result Traditionally, diets of predatory fish have of an inability to procure tissue; however, been analyzed by visual prey identification this did not meaningfully affect results, (Stevens 1966; Tucker et al. 1998; Sabal because these species were not detected. et al. 2016). While this methodology is Although Striped Bass and Sacramento logistically and economically viable, it best Pikeminnow primers were validated, describes only what prey a predator was we excluded these species as potential feeding on immediately before capture. Prey prey because we were concerned about items rapidly deteriorate in the stomach contamination from predator tissue. of predators to a point at which they

https://doi.org/10.15447/sfews.2020v18iss1art4 5 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

PCR primer sets produced single bands on an analysis samples that did not amplify with agarose gel and consistent melt curves for the universal fish primer. Soft-matrix samples species-specific products in qPCR in at least that did amplify against the universal fish duplicate for each prey species considered. primer set were then tested against a limited Voucher tissue was supplied by the UC Davis assay of species, which included Chinook Genomic Variation Laboratory, the CDFW Salmon, Steelhead, White Sturgeon (Acipenser Upper Sacramento Watershed Fisheries transmontanus), Green Sturgeon, and Pacific Project, and the field component from this Lamprey (Entosphenus tridentatus). Soft-matrix study. Voucher tissue was either frozen, samples that tested positively for one or more dried, or stored in ethanol when collected, species were conservatively counted as one and DNA was extracted using Qiagen® individual from each positive amplification. DNeasy DNA extraction kits. Table 1 contains the prey reference list and associated primers. We tested individual fish from stomach samples against targeted species primers until we The qPCR assays were run using an reached a positive result. We tested all prey fish, Eppendorf® EP realplex thermal cycler, in regardless of level of digestion, to confirm visual 96-well format. We mixed reagents using identifications. Notes on prey morphology taken 3 uL of undiluted sample DNA, 3 uL of DNase during sample processing were used to inform free water, 5 mM of primer pair, and 7.5 uL primer set selection, targeting the most likely prey of Thermo Scientific™ 2X Luminaris Color items. Occasionally, a single individual sample HiGreen qPCR Master Mix per well. All would amplify for multiple species (potentially as qPCR runs had at least two negative controls a result of inter- or intra-sample contamination), of DNase free water. We ran Brandl et al. in which case we selected the species with (2015) and Baerwald et al. (2012) primers the lowest cycle threshold value. Samples that with a qPCR program of 10 minutes at 95 °C, amplified at lower cycle thresholds were assumed followed by 40 cycles of 15 seconds at 95 °C, to contain more prey DNA. We ran a subsample and 1 minute at 60 °C. Primers designed for of amplified qPCR products on agarose gels to this study and the Jordan et al. (2010) universal validate results. Examples of positive, negative, fish primer set followed similar programs; and control amplification curves and melt profiles however, we changed the annealing step to 45 can be seen in Appendix A, Figure A1. seconds at 68 °C and 62 °C, respectively. We adjusted the annealing temperatures to maximize Data Analysis the efficiency of primers that had been designed Predator size and spatial–temporal distribution to different specifications. were analyzed using Kruskal–Wallis rank sum test and Dunn’s post hoc test of multiple Once primers had been validated, sample comparisons. We chose these non-parametric DNA was extracted from tissue using Qiagen® tests because of the non-normality of the data. Powerfecal DNA extraction kits. Powerfecal kits We analyzed predator distribution using site- are optimized for extracting DNA from low- specific catch per unit effort (CPUE) data as an quality samples high in PCR inhibitors, as is the index of abundance, with CPUE defined as the case with stomach contents. We extracted and number of fish captured per hour. Although there tested all sample DNA for DNA concentration and are inherent issues with using CPUE as a metric quality using a Thermo Scientific™ NanoDrop One for abundance, the consistency of our sampling Microvolume UV-Vis Spectrophotometer. efforts does increase the validity of its use (Haggarty and King 2006). We tested temporal Initially, we tested all samples labeled as soft distribution by season, with March through May matrix against the Jordan et al. (2010) universal classified as spring, June through August as fish primer set, to determine if samples contained summer, and September through November as any fish tissue. We excluded from further fall.

6 MARCH 2020

Table 1 Prey reference list including mitochondrial reference gene, primer sequences, gene segment length, and accession number

Common Name Segment (Latin name) Gene Primers (5’ to 3’) length Accession number American Shada FOR – TGCACGCAAACGGGGCATCA CYTB 58bp GU556214.1 (Alosa sapidissima) REV – CCTCGGCCAATGTGGGCGTAAA Chinook Salmond FOR – CCTAAAAATCGCTAATGACGCACTA CYTB 80bp KF013235 (Oncorhynchus tshawytscha) REV – GGAGTGAGCCAAAGTTTCATCAG Delta Smeltc FOR – AATGGCCAACCTTCGGAAA CYTB 90bp HQ667171 (Hypomesus transpacificus) REV – GARATATTRGAGGGTGCAGG Green Sturgeond FOR – AGGGAAAAAATGGTTAGGTCTACAGA COI 61bp KF558288 (Acipenser medirostris) REV – CCCCACTGGCGGGAAA Hardheada FOR – TGCCGGCGCAACCATCCTACA CYTB 62bp EU747218.1 (Mylopharodon conocephalus) REV – CGGCCGGGTTGTTTGATCCGGT Longfin Smeltd FOR – CTCTGCCGGGACGTCAAT CYTB 53bp KF013249 (Spirinchus thaleichthys) REV – CCCGTTAGCGTGCATATTCC Mississippi Silversidec FOR – CCGTTTGCATGCATATTTCG CYTB 73bp JN008748 (Menidia audens) REV – CCTTTTCGTCTGTTGCACACA Pacific Lampreya FOR – TTGAAGCAGGGGCTGGCACAGG COI 74bp KX389877.1 (Entosphenus tridentatus) REV – GGAGGCCCCTGTGTGGGCTAA Prickly Sculpina, e FOR – ATTGCCCTCACAGCCCTCGCAC CYTB 82bp KX353550.1 (Cottus asper) REV – TCACCAGCGGGTTAGCAGGGG Riffle Sculpina, e FOR – GGCGCCCTTTTGGGGGACGA COI 137bp JN025103.1 (Cottus gulosus) REV – GGGGCGCCGATCATTAAGGGGA River Lampreya, f FOR – CTGACTAATGTCCCACCCACCAACT CYTB 93bp KR422617.1 (Lampetra ayresi) REV – GCAGGAGAAGGAAGGTCAACTAGCA Sacramento Suckera FOR – AATCTTGCCCACGCCGGAGCC COI 132bp JN024942.1 (Catostomus occidentalis) REV – TTGAGAGATGGCTGGGGGCTTCA Sacramento Tule Percha FOR – GGGCAGAACTAAGCCAACCAGGCG COI 79bp JN026852.1 (Hysterocarpus traskii) REV – ACAAAGGCGTGGGCCGTTACAA Steelhead/Rainbow Troutd FOR – AACATAAAACCTCCAGCCATCTCT COI 59bp KF558313 (Oncorhynchus mykiss) REV – AGCACGGCTCAAACGAAAA Threadfin Shadd FOR – AAGTCCTCGGCCGATGTG CYTB 39bp KF013218 (Dorosoma petenense) REV – CATGCAAACGGAGCATCCT Western Brook Lampreya, f FOR – TCGGACGAGGAATCTACTACGGCT CYTB 118bp KY499461.1 (Lampetra richardsoni) REV – TGCCCTCATGGGAGAACGTAACCGA White Sturgeond FOR – CCCCGTTTGCATGAATGTTT CYTB 62bp KF013247 (Acipenser transmontanus) REV – CGCCCACATCTGCCGAGAT FOR – GCTTAAAACCCAAAGGACTTG Universal Fish Primerb 12S 148bp — REV – CTACACCTCGACCTGACGTT a. Benson et al. (2012). b. Jordan et al. (2010). c. Baerwald et al. (2012). d. Brandl et al. (2015). e. Primers amplified both Riffle and Prickly Sculpin DNA. f. Primers were not validated against known voucher tissue and experienced partial cross-amplification; strongest signal was selected for analysis.

https://doi.org/10.15447/sfews.2020v18iss1art4 7 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

We used cumulative prey curves, as outlined by (2) Ferry and Cailliet (1996), to determine sample size adequacy by species, habitat type, and season. With this technique, the number of new prey We chose permutational multivariate analysis of items is plotted against the number of stomachs variance (PERMANOVA) to analyze diets because analyzed, in random order. If the plot reaches an of its robustness in analyzing ecological data, as asymptote, then a sufficiently large sample size well as its ability to handle heavily zero-weighted has been obtained to describe the diet of a species data sets (Lek et al. 2011; Anderson and Walsh under the conditions studied (Ferry and Cailliet 2013; Oksanen et al. 2013). While PERMANOVA is 1996). We used the R package ‘vegan,’ function robust in analyzing data sets with heterogeneous ‘specaccum,’ to construct cumulative prey curves multivariate spread, it becomes more sensitive (Oksanen et al. 2013; Hernandez 2016; R Core to heterogeneity when groups are unbalanced Team 2018). (Anderson and Walsh 2013). Since our sample sizes of Sacramento Pikeminnow and Striped We used gravimetric values and prey Bass containing identifiable stomach contents enumerations, as determined through lab and were unbalanced (n = 30 vs. n = 47, respectively), genetic analysis, to calculate the index of relative we first tested for homogenous multivariate importance (IRI) of each prey taxon by predator spread between species groups (R package species. We used visual and genetic identifications EcoSimR::betadisper; Gotelli et al. 2015; R Core interchangeably; however, we did not include soft- Team 2018). matrix samples in IRI calculations since we could not determine prey weight for these samples. We used PERMANOVA to analyze the effect IRI is a compound value used to determine the of species, habitat type, and water temperature importance of any given prey taxon in a predator on diet composition, measured as frequency species diet (Pinkas et al. 1971; Hyslop 1980) and of occurrence. We tested a suite of other is calculated as follows: environmental and demographic variables during model construction; however, their effects on diet were insignificant, so we excluded them from IRI = (%N + %W) * %FO (1) our final model. We then subset our data (n = 20 Sacramento Pikeminnow, n = 36 Striped Bass) by where %N is total prey percent by number, %W excluding the macroinvertebrate and crayfish is total prey percent by weight, and %FO is prey groups. Using this subset data, we again total prey percent frequency of occurrence ran our model to test whether invertebrates were for a given predator group. Volumetric confounding potential differences in piscivory. measurements may be used in place of %W We ran the PERMANOVA analysis using marginal (Hyslop 1980); however, given the small testing and 10,000 permutations, which we size of prey, we chose to use gravimetric determined as sufficient to stabilize p-values. measurements because the instruments Results from all statistical tests were considered available were more precise. We used wet significant at α = 0.05. weight of prey because Glenn and Ward (1968) showed that prey wet and dry weights are highly correlated. RESULTS Size and Distribution Once we calculated IRI for each prey taxon, Over the course of the sampling period, 155 we then converted it to %IRI to increase study target species were captured, of which 68 were comparability (Cortés 1997). %IRI is calculated as Sacramento Pikeminnow and 87 were Striped follows: Bass. Of these individuals, approximately 46% of Sacramento Pikeminnow (n = 31) and 57% percent of Striped Bass (n = 50) contained

8 MARCH 2020

stomach contents (Table 2). We were unable to respectively. There was not a significant identify prey from several of the individuals difference in FL or weight between empty that contained stomach contents, which reduced and non-empty individuals for either the sample size of Sacramento Pikeminnow and Sacramento Pikeminnow (Kruskal−Wallis, FL: Striped Bass included in dietary analysis to n = 30 chi‑squared = 0.001, df = 1, p-value = 0.98; weight: and n = 47, respectively. chi‑squared = 0.001, df = 1, p = 0.98) or Striped Bass (Kruskal−Wallis, FL: chi‑squared = 0.13, Sacramento Pikeminnow were evenly distributed df = 1, p = 0.72; weight: chi‑squared = 0.17, df = 1, across all habitat types (Kruskal−Wallis, p = 0.68). Kruskal−Wallis tests showed that, for chi‑squared = 5.48, df = 2, p = 0.06), as were individuals containing identifiable stomach Striped Bass (Kruskal−Wallis, chi‑squared = 1.85, contents, Striped Bass FL was less than for df = 2, p = 0.40). Given the closeness of the test Sacramento Pikeminnow (chi‑squared = 5.27, of Sacramento Pikeminnow distribution to our df = 1, p = 0.02), while weights were not different significance threshold of alpha = 0.05, we ran (chi‑squared = 0.42, df = 1, p = 0.52). Dunn’s post Dunn’s post hoc test of multiple comparisons to hoc test confirmed that Striped Bass FL was less, test what was driving this result. Dunn’s test however, only by approximately 2.3 cm (p = 0.01). showed that CPUE of Sacramento Pikeminnow increased by 2.1 fish per hour at engineered sites Diet Composition when compared to natural sites (p = 0.02). Of the individuals that contained stomach contents, piscivory was observed in 71% of When temporal distribution was considered, Sacramento Pikeminnow and in 84% of Striped Kruskal−Wallis tests showed no difference in Bass. We were unable to identify a minimum CPUE by season for Sacramento Pikeminnow size of piscivory for either species, because the (chi‑squared = 0.37, df = 2, p = 0.83), although smallest Sacramento Pikeminnow (27 cm) and a difference was seen for Striped Bass Striped Bass (22.5 cm) that contained stomach (chi‑squared = 17.13, df = 2, p < 0.001). Dunn’s contents were both found with fish parts in their post hoc test showed this result was driven by stomachs. significantly higher CPUE of Striped Bass during summer than in fall (difference = 2.62 fish hr-1, Multivariate spread was not different between p = 0.004) or spring (difference = 4.02 fish hr-1, Sacramento Pikeminnow and Striped Bass p < 0.001). (p = 0.13), therefore meeting PERMANOVA model assumptions. Likewise, sample sizes were The average fork length (FL) and weight of determined to be adequate to describe diets by Sacramento Pikeminnow included in dietary species and habitat type, given that cumulative analysis were 35.2 cm and 0.45 kg (Table 2), prey curves for both predator species (Figure 2) and Striped Bass was 31.8 cm and 0.40 kg, and all habitat types reached an asymptote (Ferry

Table 2 Demographics of Sacramento Pikeminnow and Striped Bass that contained identifiable stomach contents. Empty rate refers to the percentage of individuals captured that did not contain any stomach contents, identifiable or otherwise.

Variable Sacramento Pikeminnow Striped Bass Number of individuals 30 47 Fork length range (cm) 27.0 − 57.0 22.5 – 47.0 Fork length mean ± SD (cm) 35.2 ± 7.4 31.8 ± 6.9 Weight range (kg) 0.20 – 1.60 0.14 - 1.00 Weight mean ± SD (kg) 0.45 ± 0.35 0.40 ± 0.24 Empty rate 54% 43% Observed onset of piscivory (cm) 27.0 22.5

https://doi.org/10.15447/sfews.2020v18iss1art4 9 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

Table 3 %IRI values for Sacramento Pikeminnow and Striped Bass captured via hook and line sampling near Chico, CA

Sacramento Prey species Pikeminnow Striped Bass American Shad 0.08 0.64 Chinook 14.57 17.03 Crayfish 2.56 0.17 Green Sturgeon 0 0.08 Hardhead 0.48 2.75 Macroinvertebrate spp. 76.9 78.09 Pacific Lamprey 0.9 0.11 Sculpin spp. 4.51 1.03 Figure 2 Species accumulation curves for Sacramento Tule Perch 0 0.1 Pikeminnow and Striped Bass. Error bars represent the possible number of new prey species added for the addition of a single randomly selected predator stomach. Species plots Table 4 PERMANOVA model testing effects of species offset for visual distinction. (Sacramento Pikeminnow and Striped Bass), habitat type, and water temperature on diets of predators captured via hook and line sampling near Chico, CA. Analysis run using and Cailliet 1996; Oksanen et al. 2013). Although frequency of occurrence as prey metric. cumulative prey curves reached an asymptote for Model: Prey Frequency of Occurrence ~ Species + Habitat Type summer and fall, they did not for fish captured + Water Temperature during the spring, precluding season as a Sum of predictor variable in our PERMANOVA model. Source df squares R2 F P Species 1 0.09 0.003 0.27 0.88 The two most important prey items for both Habitat type 2 0.43 0.02 0.66 0.72 predator species, as enumerated by %IRI, were Water temp. 1 1.04 0.04 3.22 0.02* macroinvertebrates (excluding crayfish) and Residual 72 23.17 0.93 Chinook Salmon (Sacramento Pikeminnow: Total 80 24.91 1.00 77% and 15%, respectively; Striped Bass: 78% and 17%, respectively; Table 3). PERMANOVA Significance codes: ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 modeling confirmed the similarity of diets indicated by %IRI. Prey frequency of occurrence showed no relationship with species or habitat type; however, it was significantly influenced by DISCUSSION water temperature, although it only explained Predation by native and non-native predators ~4% of the variation in diet composition (F = 3.22; in the Sacramento River system is often cited df = 1, 72; p = 0.01; Table 4). This result did not as a major factor that contributes to native change when the macroinvertebrate and crayfish species decline (NOAA Fisheries 2019), but prey groups were excluded from the PERMANOVA without sufficient information on in-river analysis. The lack of association between diet species interactions. By examining diets of the and habitat type must be qualified by the fact two important predatory fish species within the that much of the stomach contents recovered Sacramento River, this study aimed to approach were highly degraded, and prey may have been the issue of native species loss through better consumed in areas other than where predators understanding of predator diets. Although were captured. our study is only a snapshot of diets during a high-water year, it nonetheless demonstrates similarity of diets between Striped Bass and

10 MARCH 2020

Sacramento Pikeminnow in the Sacramento River as predator ambush habitat, there is no shortage (Tables 3 and 4; Figure 2). of natural structures in this section of the Sacramento River that act similarly (Whiteway Predator Diets et al. 2010). The study reach contains many %IRI and PERMANOVA modeling indicate submerged trees, or snags, which impede flow and no difference in diets between Sacramento are often targeted by recreational anglers in much Pikeminnow and Striped Bass. While there are the same way anglers target bridge pilings, for obvious life-history differences between these their ability to hold Striped Bass. Second, the two two species, on a per capita basis, neither appears water diversion facilities selected as engineered to have a higher impact on any particular prey— sampling locations—GCID and a smaller private including Chinook Salmon—than the other. pumping station—did not appear to substantially Our observed proportion of Chinook Salmon influence surface flows. This is in contrast to in predator diets was lower than was seen by other Sacramento River diversion facilities, such Thomas (1967) within the Sacramento river, as the now defunct Red Bluff Diversion Dam, and, overall, diets were substantially different which used to span the entire channel, altering than those observed within the Delta (Stevens hydraulics and increasing predation on juvenile 1966; Nobriga and Feyrer 2007). Because there salmonids by Sacramento Pikeminnow (Brown are currently no estimates of adult Sacramento and Moyle 1981). Pikeminnow or Striped Bass abundance in the Sacramento River, the total effect of predation PERMANOVA modeling showed that water on native species cannot be quantified from temperature was the only variable we measured diet composition alone. Future studies should that significantly affected predator diets. Because focus on building accurate population estimates of the association between water temperature for both Sacramento Pikeminnow and Striped and seasonality, this may indicate a temporal Bass, to clarify their role as predators and to association of predator diets, which would support quantify potential effects on prey species in the the conclusion that both Sacramento Pikeminnow Sacramento River system. and Striped Bass are opportunistically feeding on seasonally available prey populations. Had we Given the similarity of diets of Sacramento been able to capture more predators in the spring, Pikeminnow and Striped Bass, the compensatory we would have been able to directly test the effects of predator control should be considered. If association of diets with season. either Sacramento Pikeminnow or Striped Bass is resource-limited at times in the Sacramento River, Predator Distribution then their high dietary overlap suggests that Based on the results of our CPUE analysis, there control of one species would increase resources were likely more Sacramento Pikeminnow present for the other. This could potentially increase at engineered sites and more Striped Bass present the population of the species that has not been overall during summer months. Although diet did subject to control, undermining any net benefit not differ between site types, it is important to on predation. note that the increased abundance of Sacramento Pikeminnow at engineered sites may increase Predation in the Sacramento River is likely their overall predatory effect in these locations. higher near some engineered structures because Likewise, the greater abundance of Striped Bass of the favorable hydraulics that attract prey and present during the summer months may scale predators alike (Brown and Moyle 1981). However, their predatory effect on prey present within the we did not observe an association between diet Sacramento River during that time. and habitat type in our study. This can likely be attributed to two factors. First, although engineered structures such as bridge pilings do create low-water-velocity pockets, which may act

https://doi.org/10.15447/sfews.2020v18iss1art4 11 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

CONCLUSIONS REFERENCES Our study demonstrated high similarity of diets Anderson MJ, Walsh DC. 2013. PERMANOVA, between the two predator species. Although ANOSIM, and the Mantel test in the face of Sacramento Pikeminnow and Striped Bass do heterogeneous dispersions: what null hypothesis consume juveniles of native fishes such as are you testing? Ecol Monogr. [accessed 2019 Jun Chinook Salmon and Green Sturgeon, these fishes 01];83(4):557−574. https://doi.org/10.1890/12-2010.1 did not make up the majority of either species’ Baerwald MR, Schreier BM, Schumer G, May B. diet during the study period. Our results, coupled 2012. Detection of threatened Delta Smelt in the with previous diet studies, support the notion gut contents of the invasive Mississippi Silverside that Sacramento Pikeminnow and Striped Bass in the San Francisco Estuary using TaqMan exhibit prey-switching behavior, both spatially assays. Trans Am Fish Soc. [accessed 2019 Jun and temporally. This likely occurs in the presence 01];141(6):1600−1607. of high densities of certain prey, such as during https://doi.org/10.1080/00028487.2012.717521 in-river releases of hatchery Chinook Salmon. Benson DA, Cavanaugh M, Clark K, Karsch− Unfortunately, high water and turbidity did Mizrachi I, Lipman DJ, Ostell J, Sayers EW. 2012. not allow us to sample effectively when out- GenBank. Nucleic Acids Res. [accessed 2019 Jun migrating hatchery Chinook Salmon populations 01];D1(41):D36−D42. were highest. Further study should be directed https://doi.org/10.1093/nar/gks1195 at describing per capita predation by Sacramento Bonham CH. 2011. Memorandum: Striped Bass Sport Pikeminnow and Striped Bass on Chinook Salmon Fishing Regulation Amendment Proposal (December outmigrants in the Sacramento River when spring 5). Sacramento (CA): California Department of Fish flows and turbidity are low. and Wildlife. [accessed 2019 Jun 01]. Available from: ftp://ftp.dfg.ca.gov/Adult_Sturgeon_and_Striped_ Bass/signed%20memo.pdf ACKNOWLEDGEMENTS Brandl S, Schumer G, Schreier BM, Conrad JL, May B, This study was funded by the Northern California Baerwald MR. 2015. Ten real-time PCR assays for Water Association and the California Department detection of fish predation at the community level in of Fish and Wildlife (CDFW). Field crew included the San Francisco Estuary–Delta. Mol Ecol Resour. Drew Nielson and Lue Lo of California State [accessed 2019 Jun 01];15(2):278−284. University (CSU)−Chico, Brian Poxon of Pacific ht tps://doi.org/10.1111/1755-0998.12305 States Marine Fisheries Commission (PSMFC), Brown LR, Moyle PB. 1981. The impact of squawfish and Stacey Alexander of CDFW. Genetic analysis, on salmonid populations: a review. N Am J Fish statistical analysis, and methodological support Manag. [accessed 2019 Jun 01];1(2):104−111. was provided by Dr. Tyler Pilger of Fishbio, Dr. https://doi.org/10.1577/1548-8659(1981)1<104:TIOSO Andrea Schrier of University of California (UC)− S>2.0.CO;2 Davis, Michael Memeo of PSMFC, Dr. Edward Brown LR, Moyle PB. 1997. Invading species in Roualdes of CSU−Chico, and Dr. Colleen Hatfield the , California: successes, failures, and of CSU−Chico. Dr. Peter Moyle of UC−Davis and relationships with resident species. Environ Biol two anonymous reviewers provided valuable Fishes. [accessed 2019 Jun 01];49(3):271−291. feedback on an earlier version of this manuscript. https://doi.org/10.1023/A:1007381027518 Brown LR, Moyle PB. 2005. Native fish communities of the Sacramento−San Joaquin watershed, California: a history of decline. Am Fish Soc Symposium. [accessed 2019 Jun 01];45:75−98. Available from: https://www.researchgate.net/publication/287515986_ Native_fishes_of_the_Sacramento-San_Joaquin_ drainage_California_A_history_of_decline

12 MARCH 2020

[CDFG] California Department of Fish and Game. 1999. Fraser NH, Metcalfe NB, Thorpe JE. 1993. Conservation Plan for the California Department Temperature-dependent switch between diurnal of Fish and Game Striped Bass Management Plan. and nocturnal foraging in salmon. Proc R Soc Sacramento (CA): California Department of Fish and Lond. Series B: Biol Sci. [accessed 2019 Jun Game. 59 p. [accessed 2019 Jun 01]. Available from: 01];252(1334):135−139. ftp://ftp.dfg.ca.gov/Adult_Sturgeon_and_Striped_ https://doi.org/10.1098/rspb.1993.0057 Bass/Striped%20Bass%20Conservation%20Plan%20 Friesen TA, Ward DL. 1999. Management of (Part%201%20of%202).pdf Northern Pikeminnow and implications for Cavallo B, Merz J, Setka J. 2013. Effects of predator juvenile salmonid survival in the lower and flow manipulation on Chinook salmon Columbia and Snake rivers. N Am J Fish Manag. (Oncorhynchus tshawytscha) survival in an imperiled [accessed 2019 Jun 01];19(2):406−420. https://doi. estuary. Env Biol Fishes. [accessed 2019 Jun org/10.1577/1548-8675(1999)019<0406:MONPAI>2.0 01];96:393−403. .CO;2 https://doi.org/10.1007/s10641-012-9993-5 Glenn CL, Ward FJ. 1968. “Wet” weight as a method Cortés, E. 1997. A critical review of methods of for measuring stomach contents of walleyes, studying fish feeding based on analysis of stomach Stizostedion vitreum vitreum. J Fish Board Can. contents: application to elasmobranch fishes. Can J [accessed 2019 Jun 01]; 25(7):1505−1507. Fish Aquat Sci. [accessed 2019 Jun 01];54:726−738. https://doi.org/10.1139/f68-132 https://doi.org/10.1139/f96-316 Gotelli NJ, Hart EM, Ellison AM. 2015. EcoSimR: Null Deng Z, Mueller RP, Richmond MC, Johnson GE. 2010. model analysis for ecological data. version 0.1.0. Injury and mortality of juvenile salmon entrained [accessed 2019 Jun 01]. Available from: https:// in a submerged jet entering still water. N Am J cran.r-project.org/web/packages/EcoSimR/index.html Fish Manag. [accessed 2019 Jun 01];30(3):623−628. Haggarty DR, King JR. 2006. CPUE as an index of https://doi.org/10.1577/M09-153.1 relative abundance for nearshore reef fishes. Fish Durbin EG, Casas MC, Rynearson TA. 2011. Copepod Res. [accessed 2019 Jun 01];81(1):89–93. feeding and digestion rates using prey DNA https://doi.org/10.1016/j.fishres.2006.05.015 and qPCR. J Plankton Res. [accessed 2019 Jun Hernandez KM. 2016. Sex-specific diet and rockfish 01];34(1):72−82. https://doi.org/10.1093/plankt/fbr082 consumption in California sea lions (Zalophus Ferry LA, Cailliet GM. 1996. Sample size and data californianus): insights from molecular scatology analysis: are we characterizing and comparing [thesis]. [San Jose (CA)]: San Jose State University. diet properly? In: MacKinlay D, Shearer K, editors. [accessed 2019 Jun 01]; p. 1−88. Feeding ecology and nutrition in fish. Proceedings https://doi.org/10.31979/etd.49wc-7qbn of the International Congress on the Biology of Hillman TW, Griffith JS, Platts WS. 1987. Summer and Fishes, “GUTSHOP ‘96”; 1996 July 14-18, San winter habitat selection by juvenile Chinook Salmon Francisco, CA: p 71−80. [location unknown]: in a highly sedimented Idaho stream. Trans Am Fish American Fisheries Society. [accessed 2019 Jun 01]. Soc. [accessed 2019 Jun 01];116(2):185−195. https:// Available from: doi.org/10.1577/1548-8659(1987)116<185:SAWHSB> http://fishphysiology.org/wp-content/uploads/2014/02/ 2.0.CO;2 gutshop96.pdf Hunter E, Taylor N, Fox CJ, Maillard M, Taylor MI. http://fishphysiology.org/wp-content/uploads/2014/02/ 2012. Effectiveness of TaqMan probes for detection Ferr y1.pdf of fish eggs and larvae in the stomach contents of Foster JR. 1977. Pulsed gastric lavage: an a teleost predator. J Fish Biol. [accessed 2019 Jun efficient method of removing the stomach 01];81(1):320−328. contents of live fish. Prog Fish-Cult. [accessed https://doi.org/10.1111/j.1095-8649.2012.03298.x 2019 Jun 01];39(4):166−169. https://doi. Hyslop EJ. 1980. Stomach contents analysis—a org/10.1577/1548-8659(1977)39[166:PGL]2.0.CO;2 review of methods and their application. J fish biol. [accessed 2019 Jun 01];17(4):411−429. https://doi.org/10.1111/j.1095-8649.1980.tb02775.x

https://doi.org/10.15447/sfews.2020v18iss1art4 13 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

Jordan LG, Steele CA, Thorgaard GH. 2010. Universal Michel CJ, Smith JM, Demetras NJ, Huff DD, mtDNA primers for species identification of degraded Hayes SA. 2018. Non-native fish predator density bony fish samples. Mol Ecol Resour. [accessed 2019 and molecular-based diet estimates suggest differing Jun 01];10(1):225−228. impacts of predator species on juvenile salmon in https://doi.org/10.1111/j.1755-0998.2009.02739.x the , California. San Franc Estuary Kohlhorst DW. 1999. Status of Striped Bass in the Watershed Sci. [accessed 2019 Jun 01];16(4). Sacramento−San Joaquin estuary. Calif Fish Game. https://doi.org/10.15447/sfews.2018v16iss4art3 [accessed 2019 Jun 01];85(1):31−36. Available from: Morris Jr JA, Rulifson RA, Toburen LH. 2003. Life ftp://ftp.dfg.ca.gov/Adult_Sturgeon_and_Striped_ history strategies of Striped Bass, Morone saxatilis, Bass/Striped%20bass%20status%20California%20 populations inferred from otolith microchemistry. 1999.pdf Fish Res. [accessed 2019 Jun 01];62(1):53−63. Le Doux−Bloom CM. 2012. Distribution, habitat use, https://doi.org/10.1016/S0165-7836(02)00246-1 and movement patterns of sub-adult Striped Bass Moyle PB. 2002. Inland fishes of California: revised Morone saxatilis in the San Francisco Estuary and expanded. Berkeley (CA): University of Watershed, California [dissertation]. [Davis (CA)]: California Press. [accessed 2019 Jun 01]; p. 121−375 University of California, Davis. p. 34−57. Available (p. 65-153 in online version). Available from: https:// from: https://search.proquest.com/openview/ec650c31 www.waterboards.ca.gov/water_issues/programs/tmdl/ ac279b5c853549205d38f08f/1?pq-origsite=gscholar& records/state_board/1998/ref2608.pdf cbl=18750&diss=y Moyle PB, Quiñones RM, Katz JV, Weaver J. 2015. Lek E, Fairclough DV, Platell ME, Clarke KR, Fish species of special concern in California. Tweedley JR, Potter IC. 2011. To what extent Sacramento (CA): California Department of are the dietary compositions of three abundant, Fish and Wildlife. [accessed 2019 Jun 01]; 25 p. co-occurring labrid species different and related to Available from: https://nrm.dfg.ca.gov/FileHandler. latitude, habitat, body size and season?. J Fish Biol. ashx?DocumentID=104268 [accessed 2019 Jun 01];78(7):1913−1943. Nakamoto RJ, Harvey BC. 2003. Spatial, seasonal, and ht tps://doi.org/10.1111/j.1095-8649.2011.02961.x size-dependent variation in the diet of Sacramento Lindley ST, Mohr MS. 2003. Modeling the effect of Pikeminnow in the Eel River, northwestern Striped Bass (Morone saxatilis) on the population California. Calif Fish Game. [accessed 2019 Jun viability of Sacramento River winter-run Chinook 01];89(1):30−45. Available from: https://pdfs. Salmon (Oncorhynchus tshawytscha). Fish Bull. semanticscholar.org/451d/42aeaf84ea135557bdde90c [accessed 2019 Jun 01];101(2):321−331. Available 5b34447249578.pdf from: http://aquaticcommons.org/15127/ [NOAA Fisheries] National Oceanic and Atmospheric Link JS. 2002. Ecological considerations in fisheries Administration, National Marine Fisheries Service. management: when does it matter? Fisheries. 2018. NOAA Restoration Center’s Program to [accessed 2019 Jun 01];27(4):10−17. https://doi. Facilitate Implementation of Restoration Projects in org/10.1577/1548-8446(2002)027<0010:ECIFM>2.0. the Central Valley of California. NMFS No. WCR- CO;2 2017-8532. Sacramento (CA): NOAA. [accessed 2019 Loboschefsky E, Benigno G, Sommer T, Rose K, Jun 01]; 115 p. Available from: https://suscon.org/ Ginn T, Massoudieh A, Loge F. 2012. Individual-level wp-content/uploads/2018/09/2018-08-31-NOAA- and population-level historical prey demand of San Restoration-Centers-Central-Valley-BiOp.pdf Francisco Estuary Striped Bass using a bioenergetics [NOAA Fisheries] National Oceanic and Atmospheric model. San Franc Estuary Watershed Sci. [accessed Administration, National Marine Fisheries Service. 2019 Jun 01];10(1). 2019. Biological Opinion on long-term operation of https://doi.org/10.15447/sfews.2012v10iss1art3 the Central Valley Project and State Water Project. NMFS Endangered Species Act Section 7 Biological Opinion WCRO-2016-00069. Sacramento (CA): NOAA. [accessed 2019 Jun 01]; 900 p. Available from: https://www.fisheries.noaa.gov/webdam/ download/98198559

14 MARCH 2020

Nobriga ML, Feyrer F. 2007. Shallow-water piscivore− Stevens DE. 1966. Food habits of Striped Bass, Roccus prey dynamics in California’s Sacramento–San saxatilis, in the Sacramento−San Joaquin Delta. In: Joaquin Delta. San Franc Estuary Watershed Sci. Turner JL, Kelley DW, compilers. Ecological studies [accessed 2019 Jun 01];5(2). of the Sacramento–San Joaquin Delta. Part II: Fishes https://doi.org/10.15447/sfews.2007v5iss2art4 of the Delta. Fish Bulletin 136. Sacramento (CA): Nobriga ML, Feyrer F, Baxter RD. 2006. Aspects California Department of Fish Game. [accessed 2019 of Sacramento Pikeminnow biology in nearshore Jun 01];p. 68–96. Available from: ftp://ftp.dfg.ca.gov/ habitats of the Sacramento–San Joaquin Adult_Sturgeon_and_Striped_Bass/Striped%20 Delta, California. West North Am Naturalist. bass%20food%20habits%20California%201966.pdf [accessed 2019 Jun 01];66(1):106−114. https://doi. Stevens DE, Kohlhorst DW, Miller LW, Kelley DW. org/10.3398/1527-0904(2006)66[106:AOSPBI]2.0. 1985. The decline of Striped Bass in the CO;2 Sacramento−San Joaquin Estuary, California. Trans Oksanen J, Blanchet FG, Kindt R, Legendre P, Am Fish Soc. [accessed 2019 Jun 01];114(1):12−30. Minchin PR, O’Hara RB, Simpson GL, Solymos P, https://doi.org/10.1577/1548-8659(1985)114<12:TDOS Stevens MH, Wagner H, Oksanen MJ. 2013. Package BI>2.0.CO;2 ‘vegan.’ Community ecology package, version 2. Taguchi T, Miura Y, Krueger D, Sugiura S. 2014. [accessed 2019 Jun 01]. Available from: https:// Utilizing stomach content and faecal DNA analysis cran.r-project.org/web/packages/vegan/index.html techniques to assess the feeding behaviour of Pinkas L, Oliphant MS, Iverson ILK. 1971. Food habits Largemouth Bass Micropterus salmoides and Bluegill study: Food habits of albacore, Bluefin Tuna, and Lepomis macrochirus. J Fish Biol. [accessed 2019 Jun bonito in California waters. Fish Bull. [accessed 01];84(5):1271−1288. 2019 Jun 01];152:1−105. Available from: http:// ht tps://doi.org/10.1111/j f b.12341 content.cdlib.org/view?docId=kt8290062w&brand=c... Texas Instruments. 1974. Hudson River ecological here&doc.view=entire_text study in the area of Indian Point. Dallas (TX): (NY): R Core Team. 2018. R: a language and environment 1973 annual report prepared for Consolidated Edison for statistical computing. Vienna (Austria): R Company of New York. [accessed 2019 Jun 01]; Foundation for Statistical Computing. [accessed 2019 451 p. Available from: Jun 01]. Available from: http://www.R-project.org/ https://www.nrc.gov/docs/ML1002/ML100290664.pdf Rees HC, Maddison BC, Middleditch DJ, Patmore JR, Thomas JL. 1967. The diet of juvenile and adult Gough KC. 2014. The detection of aquatic Striped Bass, Roccus saxatilis, in the Sacramento− species using environmental DNA − a review of San Joaquin river system. Calif Fish Game. eDNA as a survey tool in ecology. J Applied Ecol. [accessed 2019 Jun 01];53(1):49−62. Available from: [accessed 2019 Jun 01];51(5):1450−1459. ftp://ftp.dfg.ca.gov/Adult_Sturgeon_and_Striped_ ht tps://doi.org/10.1111/1365-2664.12306 Bass/Striped%20bass%20food%20habits%20 Sabal MC, Hayes SA, Merz J, Setka J. 2016. Habitat California%201967a.pdf alterations and a nonnative predator, the Striped Bass, increase native Chinook Salmon mortality in the Central Valley, California. N Am J Fish Manag. [accessed 2019 Jun 01];36(2):309−320. https://doi.org/10.1080/02755947.2015.1121938 Sabal MC, Michel CJ, Smith JM, Hampton A, Hayes SA. 2019. Seasonal movement patterns of Striped Bass (Morone saxatilis) in their nonnative range. Estuaries Coasts. [accessed 2019 Jun 01];42(2):567−579. https://doi.org/10.1007/s12237-018-0467-7

https://doi.org/10.15447/sfews.2020v18iss1art4 15 SAN FRANCISCO ESTUARY & WATERSHED SCIENCE VOLUME 18, ISSUE 1, ARTICLE 4

Tucker ME, Williams CM, Johnson RR. 1998. Abundance, food habits and life history aspects of Sacramento Squawfish and Striped Bass at the Red Bluff Diversion Complex, including the Research Pumping Plant, Sacramento River, California, 1994−1996. Red Bluff (CA): U.S. Department of the Interior, Fish and Wildlife Service. Annual Report: Red Bluff Research Pumping Plant Report Series Volume 4. [accessed 2019 Jun 01]; 54 p. Available from: https://books.google.com/books?hl=en&lr=&id =dQNFAAAAYAAJ&oi=fnd&pg=PR6&dq=Abundan ce,+food+habits+and+life+history+aspects+of+Sacra mento+squawfish+and+Striped+Bass+at+the+Red+Bl uff+Diversion+Complex,+including+the+Research+Pu mping+Plant,+Sacramento+River,+California,+1994- 1996.+&ots=AQOyKCNZTD&sig=4Wl3Z7aPvjacsQ51I h8hyncmHic Valdez−Moreno M, Quintal−Lizama C, Gómez− Lozano R, García−Rivas MC. 2012. Monitoring an alien invasion: DNA barcoding and the identification of Lionfish and their prey on coral reefs of the Mexican Caribbean. PLoS One. [accessed 2019 Jun 01];7(6). https://doi.org/10.1371/journal.pone.0036636 Vondracek B. 1987. Digestion rates and gastric evacuation times in relation to temperature of the Sacramento Squawfish, Ptychocheilus grandis. Fish Bull. [accessed 2019 Jun 01]; 85(1):159−163. Available from: https://www.st.nmfs.noaa.gov/spo/FishBull/851/ vondracek.pdf Whiteway SL, Biron PM, Zimmermann A, Venter O, Grant JW. 2010. Do in-stream restoration structures enhance salmonid abundance? a meta- analysis. Can J Fish Aquat Sci. [accessed 2019 Jun 01];67(5):831−841. https://doi.org/10.1139/F10-021 Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, Madden TL. 2012. Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinformatics. [accessed 2019 Jun 01];13(1):134. https://doi.org/10.1186/1471-2105-13-134 Zimmerman MP, Ward DL. 1999. Index of predation on juvenile salmonids by Northern Pikeminnow in the lower Columbia River basin, 1994–1996. Trans Am Fish Soc. [accessed 2019 Jun 01];128(6):995−1007. https://doi.org/10.1577/1548-8659(1999)128<0995:IOP OJS>2.0.CO;2

16